Burner ignition detection method, device and equipment

Through image acquisition and segmentation technology, the ignition point of the burner is identified, and the detection model trained by flame samples is used to solve the problem of low accuracy of burner ignition detection, achieving higher detection accuracy and power generation safety.

CN117152675BActive Publication Date: 2024-11-05SHANTOU POWER PLANT OF HUANENG (GUANGDONG) ENERGY DEVELOPMENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310907335.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-07-21
Publication Date
2024-11-05
Estimated Expiration
2043-07-21

AI Technical Summary

Technical Problem

In the prior art, the accuracy of burner fire detection is low, which affects the safety of power generation.

Method used

The target image of the burner is collected through the image acquisition device, and the image is divided into a grid based on the preset segmentation method, which identifies the brightness, extracts the ignition point image, and inputs it to a detection model trained based on flame sample for detection, and an alarm and fire extinguishing prompt is issued.

Benefits of technology

It improves the accuracy of burner fire detection, ensures power generation safety, and prevents fire from expanding in a timely manner.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117152675B_ABST
    Figure CN117152675B_ABST
Patent Text Reader

Abstract

The present invention provides a burner ignition detection method, device and equipment, the method comprising: acquiring a target image of a burner through an image acquisition device; based on a preset segmentation method, segmenting the target image into a target number of grids; respectively identifying the brightness of each grid, and when the brightness is greater than a flame brightness threshold, determining that there is a fire point in the current grid; extracting the ignition point image based on the outline of the ignition point; inputting the ignition point image into a ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed based on training with ignition point image samples and flame samples; if the ignition point is determined to be a flame, a fire extinguishing prompt is issued to a user terminal, and an alarm is issued; if the ignition point is determined not to be a flame, the ignition point is determined to be a false alarm, and a preliminary judgment is made through brightness detection whether there is a ignition point, and then flame detection is performed through the ignition detection model, thereby ensuring the accuracy of the final ignition detection and effectively ensuring the safety of power generation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of power generation safety technology, and in particular to a burner ignition detection method, device and equipment. Background Art

[0002] The burner is a core device in a thermal power plant. It generates steam by burning flames, drives the turbine generator to rotate, and thus generates electricity. The quality of the burner directly affects the power supply capacity and efficiency of the thermal power plant, so it plays a very important role. However, during the use of the burner, fire may be observed due to burner damage. Burner fire means that the burner may have coal leakage or other faults that urgently need to be repaired, which will affect the safe production of the entire power plant.

[0003] Therefore, how to monitor the ignition status of the burner in real time has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0004] The present invention provides a burner ignition detection method, device and equipment, which are used to solve the defect of low burner ignition detection accuracy in the prior art.

[0005] The present invention provides a burner ignition detection method, comprising:

[0006] Acquire a target image of the burner by using an image acquisition device;

[0007] Based on a preset segmentation method, the target image is segmented into a target number of grids;

[0008] Identify the brightness of each of the grids respectively, and when the brightness is greater than a flame brightness threshold, determine that there is a fire point in the current grid;

[0009] Extracting a fire point image based on the outline of the fire point;

[0010] Inputting the ignition point image into a ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed by training based on ignition point image samples and flame samples;

[0011] If it is determined that the ignition point is a flame, a fire extinguishing prompt is sent to the user terminal and an alarm is issued. If it is determined that the ignition point is not a flame, it is determined that the ignition point is a false alarm.

[0012] According to a burner ignition detection method provided by the present invention, after determining that the ignition point is a flame, the method further comprises:

[0013] determining a concentration of carbon monoxide produced by the flame;

[0014] If the carbon monoxide concentration is greater than a preset carbon monoxide concentration threshold, an indoor air pollution prompt is issued.

[0015] A burner ignition detection method according to the present invention also includes:

[0016] Determining the ignition point position of the flame;

[0017] The fire extinguishing equipment is started to locate and extinguish the fire at the fire point, and the coal supply corresponding to the fire point is shut off.

[0018] According to a burner ignition detection method provided by the present invention, after respectively identifying the brightness of each of the grids, the method further includes:

[0019] Compare the brightness differences between different grids;

[0020] When the brightness difference is greater than a preset difference, it indicates that a fire point exists in the grid with greater brightness.

[0021] According to a burner ignition detection method provided by the present invention, the ignition point image is extracted based on the outline of the ignition point, comprising:

[0022] Determine the image brightness values ​​at different positions in the grid corresponding to the ignition point;

[0023] Determining the outline of the fire point based on the image brightness values ​​at the different positions;

[0024] According to the outline, the ignition point image is extracted.

[0025] A burner ignition detection method according to the present invention also includes:

[0026] Collecting video images of the ignition point within a target duration;

[0027] Identifying flame coordinates in the video image, and determining whether there is a dynamic flame inside the video image according to the flame coordinates;

[0028] If there is a dynamic flame, it is determined that a fire has occurred, otherwise it is determined to be a false alarm.

[0029] According to a burner ignition detection method provided by the present invention, after determining that the ignition point is a flame, the method further comprises:

[0030] Determining the size of the flame, and determining the degree of danger based on the size of the flame;

[0031] When the danger level is level one, an audible and light alarm will be issued;

[0032] When the danger level is level 2, a prompt is issued to stop the burner operation for maintenance;

[0033] When the danger level is level three, the burner is controlled to stop running.

[0034] A burner ignition detection method according to the present invention also includes:

[0035] Counting the number of ignitions of different parts of the burner within a preset time period;

[0036] The monitoring frequency of the different components is adjusted based on the number of fires, and the corresponding fire extinguishing methods are adjusted.

[0037] The present invention also provides a burner ignition detection device, comprising:

[0038] An acquisition module, used for acquiring a target image of the burner through an image acquisition device;

[0039] A segmentation module, used for segmenting the target image into a target number of grids based on a preset segmentation method;

[0040] A determination module, used to respectively identify the brightness of each of the grids, and when the brightness is greater than a flame brightness threshold, determine that there is a fire point in the current grid;

[0041] An extraction module, used for extracting a fire point image based on the outline of the fire point;

[0042] A detection module, used for inputting the ignition point image into a ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed by training based on ignition point image samples and flame samples;

[0043] The prompt module is used to send a fire extinguishing prompt to the user terminal and issue an alarm if it is determined that the ignition point is a flame, and to determine that the ignition point is a false alarm if it is determined that the ignition point is not a flame.

[0044] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the burner ignition detection methods described above is implemented.

[0045] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the burner ignition detection method as described in any one of the above is implemented.

[0046] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the burner ignition detection method as described above is implemented.

[0047] The present invention provides a burner fire detection method, device and equipment, the method comprising: acquiring a target image of the burner through an image acquisition device; based on a preset segmentation method, segmenting the target image into a target number of grids; respectively identifying the brightness of each grid, and when the brightness is greater than a flame brightness threshold, determining that there is a fire point in the current grid; extracting the fire point image based on the outline of the fire point; inputting the fire point image into a fire detection model to determine whether the fire point is a flame, wherein the fire detection model is pre-constructed based on training with fire point image samples and flame samples; if the fire point is determined to be a flame, a fire extinguishing prompt is issued to a user terminal, and an alarm is issued; if the fire point is determined not to be a flame, the fire point is determined to be a false alarm, and a preliminary judgment is made through brightness detection whether there is a fire point, and then flame detection is performed through the fire detection model, thereby ensuring the accuracy of the final fire detection and effectively ensuring the safety of power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 It is a flow chart of a burner ignition detection method provided by an embodiment of the present invention;

[0050] Figure 2 is a structural schematic diagram of a burner ignition detection device provided by an embodiment of the present invention;

[0051] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] Combine the following Figure 1-Figure 3 A burner ignition detection method, device and equipment of the present invention are described.

[0054] Figure 1 It is a flow chart of a burner ignition detection method provided in an embodiment of the present invention.

[0055] like Figure 1 As shown, a burner ignition detection method provided by an embodiment of the present invention mainly includes the following steps:

[0056] 101. Capture a target image of the burner through an image acquisition device.

[0057] In a specific implementation process, the image acquisition device is installed in a position where it can accurately illuminate the burner, ensuring that the image acquisition device, such as a camera, can complete the acquisition of the burner in a timely and accurate manner. The acquisition process can be real-time, and analysis and judgment are performed by acquiring the target image of the burner in real time.

[0058] The target image acquired by the image acquisition device may be a color image, a black and white image, an infrared image, or the like.

[0059] 102. Based on a preset segmentation method, the target image is segmented into a target number of grids.

[0060] After acquiring the target image, in order to more accurately identify and process the target image, you can choose to segment the target image. For example, you can divide the target image equally according to the area and divide it into 16 grids of the same size. Each grid is identified separately to ensure the accuracy of recognition.

[0061] Alternatively, the location of the fire-prone points in the target image is determined, and the target image is divided into a target number of grids according to the location of the fire-prone points, wherein the fire-prone points refer to the points where the most fires occurred in the historical period, or the high-risk fire points identified by relevant personnel. Grid division is performed by the fire-prone points, and specific analysis is performed on a case-by-case basis, which can better ensure the accuracy of image analysis and speed up image recognition.

[0062] 103. Identify the brightness of each grid respectively. When the brightness is greater than the flame brightness threshold, determine that there is a fire point in the current grid.

[0063] The brightness of the captured target image under normal working conditions is basically maintained within a certain range. If a fire occurs, the brightness of the fire point is much greater than the brightness of the unfired point. Therefore, by judging the brightness of the grid, it is possible to determine whether there is a fire point in the current grid. For example, assuming that the flame brightness threshold is a, when the brightness of the grid is less than or equal to a, it indicates that there is no fire point in the grid at this time, and when the brightness of the grid is greater than a, it indicates that there is a fire point in the grid at this time. Among them, the brightness of the grid refers to the brightness corresponding to the brightest point in the grid. It is also possible to compare the brightness differences between different grids. When the brightness difference is greater than the preset difference, it indicates that there is a fire point in the grid with greater brightness.

[0064] Through brightness recognition, it is possible to quickly identify whether there is a fire point. Not only is the principle simple, but the recognition is also accurate, which effectively improves the efficiency of fire detection.

[0065] 104. Extract the fire point image based on the outline of the fire point.

[0066] Determine whether there is a fire point in the target image by brightness judgment. If there is no fire point, no other processing is done and image acquisition is performed according to the image acquisition requirements. If it is determined that there is a fire point, further recognition processing is required to determine whether the fire point detection is accurate.

[0067] At this time, it is necessary to extract the outline of the fire point. This can be done by first determining the image brightness values ​​at different positions in the grid corresponding to the fire point. Because the brightness at different positions of the fire point is different, the outline of the fire point can be determined by judging the brightness. Based on the image brightness values ​​at different positions, the outline of the fire point is determined, that is, the boundary where the brightness changes greatly is the outline of the fire point. According to the outline, the fire point image is extracted.

[0068] 105. Input the ignition point image to the ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed based on the ignition point image samples and the flame samples for training.

[0069] The ignition point image is input into the ignition detection model, and the detection results are output, including flame and non-flame. The ignition point detection model is trained based on ignition point image samples and flame samples. Through a large number of sample experiments and tests, the accuracy of the final flame detection can be guaranteed.

[0070] The process of the fire detection model can be to identify the texture features of the fire point image. The texture features of the flame are different from the texture features of ordinary photos. Therefore, by comparing the texture features, it is possible to quickly and accurately complete the detection of whether it is a flame.

[0071] 106. If it is determined that the ignition point is a flame, a fire extinguishing reminder is sent to the user terminal and an alarm is issued. If it is determined that the ignition point is not a flame, the ignition point is determined to be a false alarm.

[0072] If the flame detection module determines that the ignition point image is a flame, a fire extinguishing prompt will be immediately sent to the user terminal, allowing the user to extinguish the fire in time and issue an alarm to ensure the safety of the burner and personnel. If it is determined that the ignition point is not a flame, the ignition point detection is determined to be a false alarm. At this time, no alarm processing is performed and normal fire monitoring can be maintained.

[0073] A burner fire detection method provided in this embodiment includes: acquiring a target image of a burner through an image acquisition device; based on a preset segmentation method, segmenting the target image into a target number of grids; identifying the brightness of each grid respectively, and when the brightness is greater than a flame brightness threshold, determining that there is a fire point in the current grid; extracting a fire point image based on the outline of the fire point; inputting the fire point image into a fire detection model to determine whether the fire point is a flame, wherein the fire detection model is pre-constructed based on training with fire point image samples and flame samples; if the fire point is determined to be a flame, a fire extinguishing prompt is issued to a user terminal, and an alarm is issued; if the fire point is determined not to be a flame, the fire point is determined to be a false alarm, and a preliminary judgment is made on whether there is a fire point through brightness detection, and then flame detection is performed through the fire detection model, thereby ensuring the accuracy of the final fire detection and effectively ensuring the safety of power generation.

[0074] Furthermore, based on the above embodiment, after determining that the ignition point is a flame, this embodiment also includes: determining the carbon monoxide concentration generated by the flame; if the carbon monoxide concentration is greater than a preset carbon monoxide concentration threshold, issuing an indoor air pollution prompt.

[0075] Specifically, after the flame is confirmed, the flame is usually caused by coal leakage, so it is very likely to produce a large amount of carbon monoxide, and it is necessary to collect the carbon monoxide concentration in time. When the carbon monoxide concentration is greater than the preset carbon monoxide concentration threshold, it indicates that there is a risk of carbon monoxide poisoning. At this time, it is necessary to issue an air pollution warning to ensure the safety of relevant personnel.

[0076] Furthermore, based on the above embodiment, this embodiment also includes: determining the ignition point of the flame; starting the fire extinguishing equipment to locate and extinguish the ignition point, and shutting off the coal supply corresponding to the ignition point.

[0077] Specifically, after confirming that a fire has occurred, it is necessary to accurately locate the location of the fire point, and then start the fire-fighting equipment at that location to locate and extinguish the fire, complete the fire-fighting in a timely manner, and shut down the corresponding coal transportation, so as to ensure the efficiency of fire-fighting and prevent the fire from further expanding and causing casualties and property losses.

[0078] Furthermore, based on the above embodiment, the present embodiment also includes: collecting video images of the ignition point within the target time length; identifying the flame coordinates in the video image, and determining whether there is a dynamic flame inside the video image according to the flame coordinates; if there is a dynamic flame, it is determined that a fire has occurred, otherwise it is determined to be a false alarm.

[0079] Specifically, in order to more accurately determine whether a fire has occurred, after determining the presence of a flame through the neural network model, a video analysis can be performed to make a more precise flame judgment. The flame will change during the combustion process. Therefore, it is necessary to detect whether the flame has changed within the target duration, that is, whether it is dynamic. If the flame is dynamic, the accuracy of the fire point can be further verified. If no dynamic flame is found, it indicates that a false detection has occurred.

[0080] Furthermore, based on the above-mentioned embodiment, after determining that the ignition point is a flame, the present embodiment also includes: determining the size of the flame, and determining the degree of danger based on the size of the flame; when the degree of danger is level one, issuing a sound and light alarm prompt; when the degree of danger is level two, issuing a prompt to stop the burner operation for maintenance; when the degree of danger is level three, controlling the burner to stop running.

[0081] Specifically, different alarms are issued according to different danger levels, which can ensure production safety and work efficiency at the same time. When the danger level is level 1, sound and light alarm prompts are issued to remind staff to extinguish the fire in time. When the danger level is level 2, maintenance prompts are issued to ensure the normal operation of the burner in subsequent work. When the danger level is level 3, the burner is controlled to stop working immediately. In this way, specific problems can be dealt with specifically, which is more in line with the needs of intelligent monitoring.

[0082] Furthermore, based on the above embodiment, the present embodiment also includes: counting the number of ignitions of different parts of the burner within a preset time period; adjusting the monitoring frequency of different parts based on the number of ignitions, and adjusting the corresponding fire extinguishing methods.

[0083] Specifically, in order to better achieve fire prevention, the number of fires of different parts of the burner can be counted, and the probability of fire of different parts can be calculated. Then, the monitoring frequency of different parts can be adjusted according to the probability of fire. For example, for parts with a higher probability of fire, real-time image information acquisition and recognition processing can be performed, and for parts with a lower probability of fire, image acquisition can be performed every five minutes. Therefore, different acquisition frequencies can be determined according to different fire probabilities, which can save resource costs. And according to the number of fires, the corresponding fire extinguishing methods can also be adjusted, including fire extinguishing by fire hydrants, fire extinguishers, and other different methods. Therefore, by adjusting the fire extinguishing method and monitoring frequency, it can not only ensure that fire detection can be achieved in a timely manner, but also ensure that fire fighting can be carried out in a timely manner when a fire occurs, and it can also save certain costs.

[0084] Based on the same general inventive concept, the present invention also protects a burner ignition detection device. The burner ignition detection device provided by the present invention is described below. The burner ignition detection device described below and the burner ignition detection method described above can be referenced to each other.

[0085] Figure 2 It is a structural schematic diagram of a burner ignition detection device provided in an embodiment of the present invention.

[0086] like Figure 2 As shown, a burner ignition detection device provided by an embodiment of the present invention includes:

[0087] The acquisition module 201 is used to acquire a target image of the burner through an image acquisition device;

[0088] A segmentation module 202 is used to segment the target image into a target number of grids based on a preset segmentation method;

[0089] The determination module 203 is used to identify the brightness of each grid respectively, and when the brightness is greater than the flame brightness threshold, it is determined that there is a fire point in the current grid;

[0090] An extraction module 204 is used to extract a fire point image based on the outline of the fire point;

[0091] The detection module 205 is used to input the ignition point image to the ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed by training based on the ignition point image samples and the flame samples;

[0092] The prompt module 206 is used to send a fire extinguishing prompt to the user terminal and issue an alarm if the ignition point is determined to be a flame, and to determine that the ignition point is a false alarm if the ignition point is determined not to be a flame.

[0093] The present embodiment provides a burner fire detection device, comprising: acquiring a target image of a burner through an image acquisition device; dividing the target image into a target number of grids based on a preset segmentation method; respectively identifying the brightness of each grid, and when the brightness is greater than a flame brightness threshold, determining that there is a fire point in the current grid; extracting a fire point image based on the outline of the fire point; inputting the fire point image into a fire detection model to determine whether the fire point is a flame, wherein the fire detection model is pre-constructed based on training of fire point image samples and flame samples; if the fire point is determined to be a flame, a fire extinguishing prompt is issued to a user terminal, and an alarm is issued; if the fire point is determined not to be a flame, the fire point is determined to be a false alarm, and a preliminary judgment is made on whether there is a fire point through brightness detection, and then flame detection is performed through the fire detection model, thereby ensuring the accuracy of the final fire detection and effectively ensuring the safety of power generation.

[0094] Furthermore, this embodiment also includes a safety detection module, which is used to:

[0095] determining a concentration of carbon monoxide produced by the flame;

[0096] If the carbon monoxide concentration is greater than a preset carbon monoxide concentration threshold, an indoor air pollution prompt is issued.

[0097] Furthermore, this embodiment also includes a positioning module, which is used to:

[0098] Determining the ignition point position of the flame;

[0099] The fire extinguishing equipment is started to locate and extinguish the fire at the fire point, and the coal supply corresponding to the fire point is shut off.

[0100] Furthermore, the determination module 203 in this embodiment is specifically used for:

[0101] Compare the brightness differences between different grids;

[0102] When the brightness difference is greater than a preset difference, it indicates that a fire point exists in the grid with greater brightness.

[0103] Furthermore, the extraction module 204 in this embodiment is specifically used for:

[0104] Determine the image brightness values ​​at different positions in the grid corresponding to the ignition point;

[0105] Determining the outline of the fire point based on the image brightness values ​​at the different positions;

[0106] According to the outline, the ignition point image is extracted.

[0107] Furthermore, this embodiment also includes a verification module, which is used to:

[0108] Collecting video images of the ignition point within a target duration;

[0109] Identifying flame coordinates in the video image, and determining whether there is a dynamic flame inside the video image according to the flame coordinates;

[0110] If there is a dynamic flame, it is determined that a fire has occurred, otherwise it is determined to be a false alarm.

[0111] Furthermore, the prompt module 206 in this embodiment is specifically used for:

[0112] Determining the size of the flame, and determining the degree of danger based on the size of the flame;

[0113] When the danger level is level one, an audible and light alarm will be issued;

[0114] When the danger level is level 2, a prompt is issued to stop the burner operation for maintenance;

[0115] When the danger level is level three, the burner is controlled to stop running.

[0116] Furthermore, this embodiment also includes a statistics module, which is used to:

[0117] Counting the number of ignitions of different parts of the burner within a preset time period;

[0118] The monitoring frequency of the different components is adjusted based on the number of fires, and the corresponding fire extinguishing methods are adjusted.

[0119] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0120] like Figure 3 As shown, the electronic device may include: a processor (processor) 310, a communication interface (Communications Interface) 320, a memory (memory) 330 and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call the logic instructions in the memory 330 to execute the burner fire detection method, which includes: acquiring a target image of the burner through an image acquisition device; based on a preset segmentation method, segmenting the target image into a target number of grids; identifying the brightness of each of the grids respectively, and when the brightness is greater than the flame brightness threshold, determining that there is a fire point in the current grid; based on the contour of the fire point, extracting the fire point image; inputting the fire point image into the fire detection model to determine whether the fire point is a flame, wherein the fire detection model is pre-constructed based on training with fire point image samples and flame samples; if the fire point is determined to be a flame, a fire extinguishing prompt is issued to the user terminal, and an alarm is issued; if the fire point is determined not to be a flame, the fire point is determined to be a false alarm.

[0121] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0122] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the burner fire detection method provided by the above methods, the method comprising: acquiring a target image of the burner through an image acquisition device; based on a preset segmentation method, dividing the target image into a target number of grids; identifying the brightness of each of the grids respectively, and when the brightness is greater than a flame brightness threshold, determining that there is a fire point in the current grid; based on the contour of the fire point, extracting the fire point image; inputting the fire point image into a fire detection model to determine whether the fire point is a flame, wherein the fire detection model is pre-constructed based on training with fire point image samples and flame samples; if the fire point is determined to be a flame, a fire extinguishing prompt is issued to a user terminal, and an alarm is issued; if the fire point is determined not to be a flame, the fire point is determined to be a false alarm.

[0123] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the burner fire detection method provided by the above-mentioned methods, the method comprising: acquiring a target image of the burner through an image acquisition device; based on a preset segmentation method, segmenting the target image into a target number of grids; identifying the brightness of each of the grids respectively, and when the brightness is greater than a flame brightness threshold, determining that there is a fire point in the current grid; extracting a fire point image based on the contour of the fire point; inputting the fire point image into a fire detection model to determine whether the fire point is a flame, wherein the fire detection model is pre-constructed based on training with fire point image samples and flame samples; if the fire point is determined to be a flame, a fire extinguishing prompt is issued to a user terminal, and an alarm is issued; if the fire point is determined not to be a flame, the fire point is determined to be a false alarm.

[0124] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0125] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A burner ignition detection method, characterized in that: include: Acquire a target image of the burner by using an image acquisition device; Based on a preset segmentation method, the target image is segmented into a target number of grids; Identify the brightness of each of the grids respectively, and when the brightness is greater than a flame brightness threshold, determine that there is a fire point in the current grid; Extracting the image of the fire point based on the outline of the fire point, including determining the image brightness values ​​at different positions in the grid corresponding to the fire point; determining the outline of the fire point based on the image brightness values ​​at different positions; extracting the image of the fire point according to the outline; Inputting the ignition point image into a ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed by training based on ignition point image samples and flame samples; If it is determined that the ignition point is a flame, a fire extinguishing prompt is sent to the user terminal, and an alarm is issued; if it is determined that the ignition point is not a flame, the ignition point is determined to be a false alarm; After determining that the ignition point is a flame, the method further includes: determining the carbon monoxide concentration generated by the flame; if the carbon monoxide concentration is greater than a preset carbon monoxide concentration threshold, issuing an indoor air pollution prompt; Determine the ignition point of the flame; start the fire extinguishing equipment to locate and extinguish the ignition point, and shut down the coal supply corresponding to the ignition point; Determining the size of the flame, and determining the degree of danger according to the size of the flame; When the danger level is level one, an audible and light alarm will be issued; When the danger level is level 2, a prompt is issued to stop the burner operation for maintenance; When the danger level is level three, controlling the burner to stop running; Counting the number of ignitions of different parts of the burner within a preset time period; The monitoring frequency of the different components is adjusted based on the number of fires, and the corresponding fire extinguishing methods are adjusted.

2. The burner ignition detection method according to claim 1, characterized in that: After respectively identifying the brightness of each of the grids, the method further includes: Compare the brightness differences between different grids; When the brightness difference is greater than a preset difference, it indicates that a fire point exists in the grid with greater brightness.

3. The burner ignition detection method according to claim 1, characterized in that: Also includes: Collecting video images of the ignition point within a target duration; Identifying flame coordinates in the video image, and determining whether there is a dynamic flame inside the video image according to the flame coordinates; If there is a dynamic flame, it is determined that a fire has occurred, otherwise it is determined to be a false alarm.

4. A burner ignition detection device, characterized in that: include: An acquisition module, used for acquiring a target image of the burner through an image acquisition device; A segmentation module, used for segmenting the target image into a target number of grids based on a preset segmentation method; A determination module, used to respectively identify the brightness of each of the grids, and when the brightness is greater than a flame brightness threshold, determine that there is a fire point in the current grid; An extraction module, used for extracting a fire point image based on the outline of the fire point; Specifically used to determine the image brightness values ​​at different positions in the grid corresponding to the fire point; determine the outline of the fire point based on the image brightness values ​​at different positions; and extract the fire point image according to the outline; A detection module, used for inputting the ignition point image into a ignition detection model to determine whether the ignition point is a flame, wherein the ignition detection model is pre-constructed by training based on ignition point image samples and flame samples; A prompt module is used to send a fire extinguishing prompt to the user terminal and issue an alarm if the ignition point is determined to be a flame, and to determine that the ignition point is a false alarm if the ignition point is determined not to be a flame; determine the carbon monoxide concentration generated by the flame; and issue an indoor air pollution prompt if the carbon monoxide concentration is greater than a preset carbon monoxide concentration threshold; A fire extinguishing module is used to determine the ignition point of the flame; start the fire extinguishing equipment to locate and extinguish the ignition point, and shut down the coal supply corresponding to the ignition point; An alarm module is used to determine the size of the flame and determine the degree of danger according to the size of the flame; when the degree of danger is level one, a sound and light alarm prompt is issued; when the degree of danger is level two, a burner stop maintenance prompt is issued; when the degree of danger is level three, the burner is controlled to stop running; The adjustment module is used to count the number of ignitions of different components of the burner within a preset time period; adjust the monitoring frequency of the different components based on the number of ignitions, and adjust the corresponding fire extinguishing methods.

5. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the burner ignition detection method according to any one of claims 1 to 3 is implemented.